The 13-week cash flow forecast

It is the tool restructuring specialists reach for first, which has given it a slightly grim reputation. That reputation is unearned. The same discipline is what lets a healthy business commit to a hire, a piece of equipment or a slower-paying but larger customer with confidence rather than hope, and it is the single most useful thing a business can add to its financial routine.

Why the horizon is thirteen weeks

Thirteen weeks is one quarter. That length is not arbitrary; it sits at the point where two opposing problems balance out.

Shorter horizons are accurate and useless. A four-week view tells you what you already suspect, and by the time a shortfall appears in it your options have narrowed to borrowing at short notice or not paying someone. Longer horizons contain real information but stop being knowable at a weekly level. Beyond a quarter you are forecasting customers you have not yet invoiced, and the weekly detail becomes theater.

A quarter also lines up with the rhythms a business already runs on: quarterly tax obligations, most debt covenant reporting cycles, seasonal swings and the normal length of a collections cycle. A shortfall that appears in week nine gives you eight weeks of options, and eight weeks is enough to accelerate collections, defer a discretionary payment, reschedule an order, draw on a facility on sensible terms, or simply decide not to do the thing that caused it.

Why it is not your budget, and not your profit and loss statement

The reason a separate document exists is that accrual accounting deliberately answers a different question.

Your income statement records revenue when it is earned and costs when they are incurred. That is the correct way to measure performance, and it is exactly the wrong way to see whether Friday's payroll clears. A business can invoice a record month, report a strong profit, and still be unable to pay its people, because the invoice has sixty-day terms and payroll has none. This is the gap that catches most growing businesses: growth consumes cash before it produces it, and the profit and loss statement reports the good news several weeks before the bank account confirms it.

A 13-week forecast uses the direct method, which means it tracks the actual movement of money rather than adjusting profit for non-cash items. Every line is a payment somebody makes or receives on a date. That is why it can show the timing gap that a budget structurally cannot. The broader case for forecasting cash separately is set out in what cash flow forecasting is.

The structure, line by line

The layout is always the same: thirteen columns, one per week, with the closing balance of each week becoming the opening balance of the next.

Keeping facility headroom on its own row matters more than it sounds. A closing balance of $40,000 means something very different with an untouched $250,000 line behind it than it does without one, and a forecast that hides the difference will cause the wrong decision.

How to build one

1. Fix the starting balance. Use the reconciled bank balance today, across all accounts, net of checks and payments already in flight. Everything downstream inherits this number, so an unreconciled start makes the whole schedule wrong by a constant.
2. Forecast receipts from behavior, not from terms. Go through the receivables ledger customer by customer and place each invoice in the week that customer actually pays. If a large account has averaged fifty-two days for the past year, the invoice belongs in week eight, not in week five because the terms say thirty days. This single adjustment is where most forecasts stop being fiction.
3. Lay out committed disbursements. Payroll dates, the payables ledger with its due dates, rent, debt service, insurance renewals, quarterly tax deadlines and any deposit already promised. These are known, and getting them onto the right week is clerical rather than analytical.
4. Add the variable and the discretionary separately. Materials and subcontractors that scale with the work you have booked, then discretionary spending as its own block. Keeping the discretionary items visible on their own line is what lets you see, in week nine, exactly which levers you still hold.
5. Net it out and read the trough. The number that matters is not the closing balance in week thirteen. It is the lowest balance at any point in the quarter, and the week it happens.
6. Roll it forward every week. Replace week one with what actually happened, add a new week thirteen at the far end, and compare the forecast against the actual before you do anything else.

The roll-forward is the whole point

A 13-week forecast built once and filed is an expensive spreadsheet. Built weekly, it becomes something else, because the weekly comparison of forecast against actual is what teaches the business how it really behaves.

The variance review takes fifteen minutes and asks three questions. Which receipts arrived late, and is that customer now a pattern rather than an exception? Which disbursements were larger than expected, and was that a one-off or a standing underestimate? And what has changed in the last week that moves the trough earlier or deeper?

After six or eight cycles the forecast stops being a guess. The collection assumptions are calibrated to real behavior, the recurring costs are right, and the owner can trust week six enough to make a decision on it. That is the state worth getting to, and it is reachable inside two months of consistent effort.

Where forecasts go wrong

• Using invoice terms instead of payment history. The most common error, and the one that makes an entire forecast optimistic by a fixed number of weeks.
• Starting from an unreconciled balance. Every week inherits the error.
• Building it monthly. A month is long enough to hide a two-week trough entirely. Payroll does not average out.
• Forecasting revenue instead of collections. A signed contract is not cash, and neither is an invoice.
• Omitting the irregular items. Quarterly taxes, annual insurance renewals, bonus runs and the extra payroll in a three-payroll month are the classic surprises, and all of them are on a calendar somewhere.
• Never comparing against actuals. Without the variance review the forecast never improves, and a forecast that never improves eventually gets ignored, which is worse than not having one.
• Treating it as an accounting deliverable. It is a decision tool. If nobody changes a decision because of it, it is not being used.

Who should be running one

Not every business needs weekly cash visibility. The ones that do share a common feature, which is that the cash balance moves faster than the reporting does.

That includes businesses with long receivable cycles or significant work in progress, where money goes out weeks before it comes back. Seasonal businesses, where a good year still contains a difficult quarter. Companies carrying debt covenants, where the consequence of a miss is not just inconvenience. Businesses growing quickly, since growth consumes working capital. Anyone integrating an acquisition. And any owner who has looked at a profitable month and a thin bank balance and not been able to explain the difference.

If that last one sounds familiar, the underlying gap between accounting profit and money in the bank is worth understanding on its own terms, and we work through it in profit and cash flow targeting.

Having someone else run it

Most owners can build a first version of this in an afternoon. The part that fails is not the build, it is the fifty-two weeks of roll-forward afterwards, because it lands on whoever is busiest and gets skipped in exactly the weeks it matters most.

Running the forecast, keeping the collection assumptions honest, and sitting down with the owner monthly to turn the trough into decisions is a standing part of what we do. It is described in full under cash flow management and forecasting, and it sits inside the broader fractional CFO service.

Consulting is billed at $250 per hour, and ongoing work runs on monthly advisory plans priced to the needs of the business. The first conversation is a free thirty-minute call and you can book one on our calendar.